Using data to describe the root causes of adolescent pregnancy and school dropout

In March, Stephanie Psaki of the Population Council asked: “Does getting pregnant cause girls to drop out of school?” If teen pregnancy and dropout rates are high, which causes which? Her question has far-reaching implications for international development interventions for adolescent girls. Psaki argues that the answer depends largely on context. In an area where girls have access to quality education and opportunities for employment, and where girls and their parents value education (characterized as “Country A” by Psaki), getting pregnant may be the sole reason for school dropout and result in the economic disempowerment of women. However, in “Country B,” the situation is more complex. There are fewer opportunities for girls, and parents and communities place little value on education for girls. Girls drop out of school for a host of reasons, and get married and/or pregnant because of a lack of other life options. In short, as Psaki stated: “The solution to the problem of schoolgirl pregnancy — and to school dropout more broadly — is not one-size-fits-all.” Despite similar rates of early pregnancy, different contexts may have different needs and require different interventions to create change. Recently, I led a digital data collection initiative in two districts in Indonesia to prepare for a Plan International program addressing adolescent reproductive health and life skills. Findings from our data collection across these two islands provide a nearly perfect real-world example of Psaki’s theoretical Country A and Country B. We used the “Girl Roster,” a digital mapping tool developed by the Population Council to collect data in our target districts of Dompu and Sikka. The Girl Roster measures five basic, nonsensitive indicators: a girl’s age, education status, marital/motherhood status and living situation (with or without parents). This tool differs from many digital data collection methods in that it covers all the girls in a community rather than a sub-sample; it is very short (just 10 minutes per household); and it is conducted by front-line field staff rather than external enumerators. We found that, while the rates of early pregnancy were roughly the same in Dompu and Sikka, the broader context and the rates of school dropout differed dramatically across the two islands. Among young women aged 18-20, 26 percent in Dompu and 20 percent in Sikka were already married. Early pregnancy was slightly less common, with 20 percent and 14 percent of young women having already given birth in Dompu and Sikka respectively. However, the education status of girls was a different story. While premature dropout was uncommon in Dompu district, the majority of girls in Sikka district had dropped out of school before they reached high school. Given the geography of Sikka, this is not surprising. The villages where we collected data were located in the mountains, several hours by bus from the nearest high school. In order to go to high school at all, a girl would have to find money to attend boarding school and leave her family. For a village of rice farmers, this seems like an absurd luxury. In Dompu, however, the land is flat and easy to navigate, and high schools are close by. Opportunities for livelihoods are diverse and abundant, and, although rates of sexual activity were not measured, girls were far more open about their sexuality. Girls freely bragged about having many boyfriends and dreamed of going to college and becoming social workers, teachers or police officers. In contrast, when I asked a teenage girl in Sikka what she wanted to do after she finished school, she looked at me somewhat blankly. “I haven’t thought about it before,” she said. “I think I’ll work on a farm and help my parents.” Parents in Sikka gave various reasons for their children dropping out of school, citing financial reasons or the distance to school. One Plan staff member noted, “Parents think that education isn’t good for children; it just keeps them from making money. There is no motivation to send children to school.” In Dompu, parents and community members were clear about what they perceived to be the cause of dropout. One village leader commented, “We have had programs in this village to prevent early marriage, and they have been quite successful. The problem is that girls get pregnant before they get married and drop out of school. There’s nothing we can do about that.” Essentially, Dompu was a quintessential Country A, with Sikka as Country B. So what now? The different contexts of adolescent pregnancy in Dompu and Sikka necessitates different interventions; while improved access to contraception might be enough to keep girls in school in Dompu, a program in Sikka would need to address the broader context of girls’ lives. Being an adolescent girl means constantly making trade-offs and seeking permission. Access to condoms does not mean she can successfully negotiate their use. Access to school does not mean she will be spared from housework and field labor. Even if she graduates from school, she may still be unable to access community resources without getting married. Getting an education might make her more “marriageable,” or it could make her practically unmarriageable in her village. Beyond the question of whether pregnancy causes dropout, or vice-versa, we must address the complicated environment that an adolescent girl must navigate. To do so, our programs for adolescent girls must be evidence-based and holistic. Without an empirical understanding of the target community context, the program may be unsustainable at best and harmful at worst. And the intervention package must consider the whole girl within the structural context of her community, building a combination of personal, social and financial assets, so that she has the power to make her own decisions about her body and future. Join the Devex community and access more in-depth analysis, breaking news and business advice — and a host of other services — on international development, humanitarian aid and global health.